The fast increase in volume and speed of information created by mobile devices, along with the availability of web-based applications, has considerably contributed to the massive collection of data. Approximate Nearest Neighbor (ANN) is essential in big size databases for comparison search to offer the nearest neighbor of a given query in the field of computer vision and pattern recognition. Many hashing algorithms have been developed to improve data management and retrieval accuracy in huge databases. However, none of these algorithms took bandwidth into consideration, which is a significant aspect in information retrieval and pattern recognition. As a result, our work created a Geo-SPEBH algorithm to solve this basic gap. The paper then assesses the performance of the Geo-SPEBH algorithm in terms of bandwidth in a distributed computing environment. Geo-performance SPEBH's was compared to existing state-of-the-art approaches using a network analyzer called Wire shark. The simulation results reveal that during retrieval, the same kb/sec of data is carried from source to destination and from destination to user. When the coding length is 8bit, the findings show that 0.091kb/sec of data is required to transport data from source to destination. Each algorithm with the same bit code requires the same amount of bandwidth to convey data.
Othman, Abubakar Usman; Timothy, Moses; Umar, Aisha Yahaya; Audu, Abdullahi Salihu; Souley, Boukari; and Gital, Abdulsalam Ya’u
"Evaluation of Geo-SPEBH algorithm based on Bandwidth for Big Data retrieval in Cloud Computing,"
Al-Bahir Journal for Engineering and Pure Sciences: Vol. 1
, Article 4.
Available at: https://doi.org/10.55810/2312-5721.1005